In this PSM based control study, patients with a prior KA had a higher risk of reoperation, revision, stiffness and PJI compared with controls following subsequent TKA. Male and prior KA for ACL injury were independent risk factors of postoperative severe complications in patients with prior KA. The time dependent effect analysis suggested it was more reliable to perform a TKA at least 9 months after the prior KA.
Our result presented inferior clinical outcomes in patients with prior KA, which was consistent with the most recent study. Alex et al. utilized the Humana insurance database to review 3357 TKAs with a prior KA and 134,662 controls. The multivariate analysis suggested the prior KA was associated with higher prevalence of revision, postoperative stiffness and PJI . However, Viste et al. reported conflicting results . They retrospective reviewed a single institutional database including 160 TKAs had a prior KA. They matched a control cohort at a 1:2 ratio and compared Knee Society Score (KSS), ROM, complications and survivorships with a mean followup of 9 years. They found the clinical outcomes of TKAs with prior KA were comparable with that of controls.
To be our best knowledge, although several studies have evaluated the effect of prior KA on TKA, there was no study to identify risk factors of worse outcomes in these patients. The present study found male and KA for PCL injury were associated with postoperative complications in patients with prior KA. Male patients had worse outcomes as male may be more active than female. Several studies have suggested TKA after ACL reconstruction resulted in worse outcomes following TKA. Watters et al. reviewed 122 patients with prior ACL reconstruction with a minimum of 2-year follow-up. They indicated TKA with a prior ACL reconstruction had a higher risk of longer operative time and early reoperation . Chong et al. performed a retrospective study including 101 cases with prior ACL reconstruction and 202 controls . However, they found There was no statistical difference in estimated blood loss and postoperative complication between ACL group and controls.
It’s critical to explore the time dependent effect of prior KA on the subsequent TKA to determine the timing of TKA. The optimal time to perform a TKA after KA was controversial in the literature. A study by Piedade et al. reviewed 60 primary TKA with a prior KA and 1,119 controls, and no time-dependent effect was found . However, both Werner et al. and Barton et al. recently reported patients who underwent TKA within 6 months after KA had worse Patient Reported Outcome and higher risk of postoperative complications [7, 8]. The most potential limitation of the two studies was to determine the cutoff of time arbitrarily. Considering time from KA to TKA as a continuous variable, the present study created a smoothing spline plots and conducted the two-piecewise linear regression analysis to explore the timing of TKA. We found patients who were scheduled to undergo TKA should wait at least 9 months after KA.
There are several limitations to the present study. First, the study design was retrospective in nature and thus was subject to its inherent biases, such as a recall bias. Second, although we tried our best to identify patients with prior KA through medical records and institutional database, we may miss several cases. Third, the sample size may be inadequate, and the possibility of a type-II error exists. Fourth, we did not analyze Patient Reported Outcome as only HSS score was available before 2018 in my institution.